A current source can be easily made using a conventional bipolar transistor as shown in FIG. 1 where the base voltage is fixed by a voltage source V.sub.1. The output current over the operating voltage range of the transistor is fairly constant and can be calculated from: ##EQU1##
The current source may be switched by inserting a differential transistor pair in series with the output current as shown in FIG. 2. Depending on the polarity of the switching signal input, the current from Q1 can be diverted through either Q2 or Q3. A current of opposite polarity can be produced by substituting PNP transistors for the NPN types shown in FIGS. 1 and 2. By using both types connected as in FIG. 3, a bi-directional current can be fed to the load according to the state of the input switch signals.
A problem exists when using this arrangement on an integrated circuit process because PNP transistors may not be available or are very often limited in gain and frequency response preventing very rapid switching of the current.
A circuit similar to that in FIG. 3 is often used as a charge pump circuit in known phase locked loop frequency synthesisers where up and down signals from the phase detector drive the appropriate +ve or -ve switch inputs to provide a correction signal to the voltage controlled oscillator to maintain phase lock. The system is shown in FIG. 4.
In operation the phase detector provides a continuous stream of pulses of the appropriate polarity to maintain the divided VCO signal in phase with the reference input to the phase detector. The pulses applied to OP AMP vary in width and polarity according to the phase error but are always of constant amplitude. This circuit is generally known as a charge pump circuit.